Improvement of IgA Nephropathy and Kidney Regeneration by Functionalized Hyaluronic Acid and Gelatin Hydrogel

Abstract

Background:

Immunoglobulin A (IgA) nephropathy (IgAN) is one of an important cause of progressive kidney disease and occurs when IgA settles in the kidney resulted in disrupts kidney’s ability to filter waste and excess water. Hydrogels are promising material for medical applications owing to their excellent adaptability and filling ability. Herein, we proposed a hyaluronic acid/gelatin (CHO-HA/Gel-NH₂) bioactive hydrogel as a cell carrier for therapeutic kidney regeneration in IgAN.

Methods:

CHO-HA/Gel-NH₂ hydrogel was fabricated by Schiff-base reaction without any additional crosslinking agents. The hydrogel concentrations and ratios were evaluated to enhance adequate mechanical properties and biocompatibility for further in vivo study. High serum IgA ddY mice kidneys were treated with human urine-derived renal progenitor cells encapsulated in the hydrogel to investigate the improvement of IgA nephropathy and kidney regeneration.

Results:

The stiffness of the hydrogel was significantly enhanced and could be modulated by altering the concentrations and ratios of hydrogel. CHO-HA/Gel-NH₂ at a ratio of 3/7 provided a promising milieu for cells viability and cells proliferation. From week four onwards, there was a significant reduction in blood urea nitrogen and serum creatinine level in Cell/Gel group, as well as well-organized glomeruli and tubules. Moreover, the expression of pro-inflammatory and pro-fibrotic molecules significantly decreased in the Gel/Cell group, whereas anti-inflammatory gene expression was elevated compared to the Cell group.

Conclusion:

Based on in vivo studies, the renal regenerative ability of the progenitor cells could be further increased by this hydrogel system.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13770-022-00442-8.

Introduction

Kidney disease has become a growing concern globally. The ability to regenerate injured renal tissue is critical for retarding the progression to chronic kidney disease (CKD), and eventually to end-stage renal disease. Immunoglobulin A (IgA) nephropathy (IgAN), the most common form of glomerular disease, is a leading cause of CKD and renal failure [1]. It is defined by the deposition of IgA in the glomerular mesangium and is closely related to lymphocyte dysregulation and undesirable immune responses [2].

However, the cause of primary IgAN and the mechanism underlying mesangial IgA deposition in IgAN are unclear; thus, effective treatment with IgAN is currently not available. However, stem cell or progenitor cell therapy is considered a potential therapeutic strategy for tissue regeneration [3]. Therefore, recent advances in research have proposed using stem cells for kidney regeneration, even though the kidney is a delicate organ with a limited regenerative capacity compared to other organs [4].

Considerable interest in regenerating damaged tissue has been initiated using stem cells as a therapeutic option [5]. One of the most important features of stem cells is their capacity to divide and induce more stem cells (self-renewal potency and clonogenicity) or more differentiated precursors, which is associated with their potential to differentiate into different particular cell types [6]. Nevertheless, not all stem cell types have the same differentiative and therapeutic potential. Renal progenitor cells (RPCs) are promising candidates for the experimental evaluation of CKD because they can be easily amplified, maintained in culture, and induced to differentiate into tubular cells [7–9].

One approach to enhancing and prolonging the cellular functions for therapeutic purposes is to encapsulate cells within a hydrogel that replicates the supportive functions of the extracellular matrix (ECM). Injectable hydrogels have been extensively studied for pharmaceutical and regenerative medicine applications [10]. Various physical and chemical cross-linking techniques have been utilized to fabricate biodegradable hydrogels, including Diels–Alder “Click” cross-linking, enzyme cross-linking, Michael addition, and ionic interaction [11].

The Schiff-base reaction has been utilized to prepare hydrogels and provides rapid gelation [12–14]. The Schiff-base linkage is in rapid equilibrium (association/dissociation) at neutral pH and provides outstanding injectability and self-healing properties under physiological conditions [15]. Because of their superior biocompatibility and biodegradability, polysaccharides such as starch, dextran, chondroitin sulfate and hyaluronic acid have been utilized to create hydrogels for drug delivery, cells carriers and wound healing [16, 17].

Among these polymers, hyaluronic acid (HA) has been widely exploited in tissue engineering applications due to its outstanding biocompatibility and biodegradability as well as fascinating viscoelastic property [18–20]. However, HA is highly soluble and often displays poor mechanical properties and rapid degradation in vivo. Thus, HA has been modified chemically or by cross-linking to improve its mechanical properties, viscosity, solubility, degradation, and biological properties [20]. Gelatin is a biocompatible and biodegradable biomaterial, the major protein component of natural ECM. Gelatin is presumed to retain some of the information signals, such as the arginyl-glycyl-aspartic acid sequence of collagen, enhancing cells adhesion and proliferation [21, 22].

In this study, we enhanced the performance of natural hydrogels using the Schiff-based cross-linking of HA with gelatin for therapeutic kidney regeneration in IgAN (Fig. 1). Such studies have not been previously conducted to the best of our knowledge. A hydrogel of HA was generated by the Schiff-base reaction between the aldehyde groups of HA and the amine groups of gelatins without using any additional cross-linking agents. HA and gelatin hydrogel systems as bioactive materials in tissue engineering have been reported [23]. HA and gelatin promoted cell differentiation by modulating hydrogel process with controlled Gel-HA ratio [24]. In addition, hybrid hydrogel was fabricated based on HA and gelatin for use as bioinks in tissue regeneration [25]. Moreover, our previous study showed that HA and gelatin hydrogel could decrease capsular tissue responses with looser collagen distribution and reduced cytokine expression on the hydrogel-coated poly(dimethylsiloxane) for implantable medical device-induced fibrosis [26]. Herein, gelatin with HA has been cross-linked with a controlled Gel-HA ratio to modulate the injectable hydrogel. HA increases matrix stiffness and gelatin promotes cell adhesion. The equilibrium swelling ratio, morphology, microstructure, and rheological properties were investigated for further in vitro and in vivo biocompatibility studies. The hyaluronic acid /gelatin (CHO-HA/Gel-NH₂) hydrogel was assessed for biocompatibility and renal therapeutic efficacy. Human urine-derived renal progenitor cells (URPCs) were introduced into the hydrogels and subsequently injected into high serum IgA ddY (HIGA) mice kidneys. We attempted cell therapy to minimize pathophysiological indicators, where the hydrogel serving as a bioartificial niche for enhancing cell and tissue regeneration. In addition, the biocompatibility and therapeutic potential of the hydrogel as a carrier were analyzed.

Fig. 1.

Preparation of hyaluronic acid/gelatin (CHO-HA/Gel-NH₂) hydrogel via Schiff-base reaction and injection of human urine-derived renal progenitor cells (URPC) encapsulated hydrogel into the renal cortex for improvement of immunoglobulin A (IgA) nephropathy

Materials and methods

Materials

Sodium hyaluronate (medium molecular weight, 1,000,000 − 1,600,000 Da) was purchased from BioScience (Dummer, Germany). Gelatin powder type A (~ 175 Bloom) was acquired from Electron Microscopy Science (Hatfield, PA, USA). Sodium periodate (NaIO₄; ACS reagent; ≥ 99.8%), ethylenediamine Reagent Plus® (≥ 99%), and anhydrous ethyleneglycol (≥ 99.8%) were obtained from Sigma-Aldrich. Carbosynth (Newbury, UK) supplied 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC). All of the chemicals were used without further purification. Hyclone phosphate buffer saline (PBS) was purchased from Thermo Fisher Scientific (Waltham, MA, USA) as were the Spectra/Por 3 dialysis membranes (MWCO 3,500 and 12,000 − 14,000).

Preparation of aldehyde hyaluronic acid (CHO-HA)

Aldehyde hyaluronic acid (CHO-HA) containing aldehyde groups was prepared by the reaction between HA and sodium periodate (NaIO₄) [27–30]. Briefly, 1 g of HA was dissolved in 100 mL distilled water at a concentration of 10 mg mL⁻¹. A certain concentration of NaIO₄ was added to HA solution. The reaction in aqueous medium was continued for 24 h at room temperature in the dark under magnetic stirring. Ethylene glycol was then added to inactivate the unreacted sodium periodate. Thereafter, the resultant solution was dialyzed (MWCO 12,000 − 14,000, Spectra/Por membrane) exhaustively for 3 days against double distilled water (DDW) and lyophilized to obtain the CHO-HA product.

Preparation of gelatin amine (Gel-NH₂)

Gelatin was dissolved in PBS to a final concentration of 5 wt% at 37 °C. Excess ethylenediamine was added to the gelatin solution and reacted for 5 min [31]. Thereafter, the pH of the solution was adjusted to 5.0 by adding 5 M of hydrochloric acid (HCl), followed by a certain amount of EDC. The reaction mixture was stirred at 37 °C for 18 h and then dialyzed (MWCO 3500, Spectra/Por membrane) against DDW for 48 h to remove the excess ethylenediamine and EDC. The dialyzed solution was freeze-dried at − 80 °C to obtain modified gelatin.

Characterization of CHO-HA and Gel-NH₂

Fourier transform infrared (FTIR) analysis of CHO-HA and Gel-NH₂ was performed to confirm the expected pendant functionalities. ¹H-NMR spectra were acquired at room temperature with a Varian 600 NMR spectrometer using D₂O as the solvent. Sample concentrations of 1 − 50 mg mL⁻¹ produced clear signals, provided the sample was completely dissolved. The sample solution (1 mL) was transferred to 5 mm NMR tubes and tetramethylsilane (TMS) was used as an internal reference.

Fabrication of hydrogel and measurement of gelation time

In a glass test tube, 300 μl of 10% CHO-HA solution was mixed with 700 μl of 20% Gel-NH₂ solution at 37 °C by magnetic stirring at 200 rpm. The time required for the magnetic bar to stop spinning was recorded as the gelation time of the mixed solution.

Equilibrium swelling ratio and degradation

The swelling and degradation of the hydrogels was evaluated gravimetrically. The swelling temperature was set as 37 °C and the pH was maintained at 7.4 using PBS. At predetermined time intervals, the hydrogel samples were removed from the buffer solution and weighed after blotting the excess water from the surface with a filter paper. The equilibrium swelling was established as equilibrium swelling (%) = $\left[\frac{Ws-Wi}{\mathit{Wi}}\times 100\right]$, where W_i is the weight of the initial hydrogel and W_s is the equilibrium weight of the swollen hydrogel [32].

Morphology

The morphology of the hydrogel was observed using a scanning electron microscope (SEM; S-3400 N, Hitachi, Tokyo, Japan). The hydrogel samples were freeze-dried and subsequently lyophilized. The samples were sectioned using a scalpel, and the cross-sections were coated by sputtering. All micrographs were the product of secondary electron imaging and were used for evaluation of the surface morphology at different magnifications using an accelerating voltage of 5 kV.

Rheological properties

A Rheometer (Model ARES-G2) instrument with parallel plate geometry was used to evaluate the viscoelastic behavior of hydrogel, and oscillation frequency sweep tests were performed. Briefly, CHO-HA 10% solution and Gel-NH₂ 20% were mixed at pH 7.4 and incubated at 37 °C for 2 h to obtain constructed of hydrogel before analysis. The hydrogels were procured on the parallel plated at 25 °C. A constant amplitude $\left(\gamma =1\%\right)$ was applied to the hydrogel at frequency between of 1 − 10 rad s⁻¹. The storage modulus (G′) and loss modulus (G″) represent the elastic properties of materials and viscous character of hydrogel respectively.

Generation and culture of iNPCs

Human iNPCs generated from human urine-derived cells (UCs) were kindly provided by cell function regulation laboratory, department of biotechnology, college of life sciences and biotechnology, Korea University [33]. Briefly, the transduced UCs with key pluripotency transcription factors were immediately exposed to the nephron progenitor niche to maintain and expand undifferentiated NPCs, including recombinant human FGF9, heparin, bone morphogenetic protein 7 (BMP7), ALK inhibitor (LDN-193189), CHIR99021 inhibitor, and ROCK inhibitor Y-27632. This signaling environment was created by modifying the previously described nephron progenitor expansion medium, in which renal progenitors from embryonic kidneys or human embryonic stem cells could be propagated by manipulating the BMP, fibroblast growth factor, and WNT signaling pathways while maintaining epithelial differentiation potential. This strategy focuses on driving cells toward an epigenetically unstable/plastic state during reprogramming, which allows them to be directed to an alternative cell fate, by passing the pluripotent state, when exposed to proper signaling environments. At 12 days post-transduction, cell colonies were observed and expanded as single cells on Matrigel-coated plates by manual picking. Upon further passaging, putative iNPCs exhibiting a small, spindle-shaped mesenchymal morphology were observed. The iNPCs were cultured on a Matrigel-coated plate in serum-free medium (Advanced RPMI, Corning, Gibco/Invitrogen, catalog number:12633012) supplemented with 100 ng mL⁻¹ FGF9 (PeproTech, Cranbury, NJ, USA, catalog number:100–23), 30 ng mL⁻¹ BMP7 (Peprotech, catalog number:123–03), 1.25 μM CHIR99021 (R&D Systems, catalog number: 4423/50), and antibiotics in a humidified incubator with 5% CO₂ in air at 37 °C. The medium was changed on alternate days.

Cell viability and cell proliferation in vitro

A live/dead assay (Live/Dead kit: 0.1 mM calcein AM, 0.4 mM ethidium bromide homodimer-1; Invitrogen) was used to determine whether the hydrogel created a microenvironment suitable for cell culture in a humidified cell incubator with 5% CO₂ at 37 °C. Briefly, the iNPCs (p = 7, 1 × 10⁵ cells) were suspended in 70 μL of Gel-NH₂ solution and placed in a 24-well plate. Thirty microliters of CHO-HA were added to the Gel-NH₂ solution and mixed well to form a hydrogel. The hydrogels were incubated for 15–30 min to complete cross-linking, and 500 μL of medium was then added. The iNPC-constructed hydrogels were washed with PBS and stained with the Live/Dead kit at the selected time interval. After 30 min, the samples were visualized using a fluorescence microscope. All experiments were performed in triplicate. According to the manufacturer’s instructions, cell proliferation was quantified on days 1, 3, and 7 using a Cell Counting Kit-8 assay (CCK-8; Dojindo Laboratories, Kumamoto, Japan). Briefly, the iNPC-encapsulated hydrogel (n = 4) was gently rinsed in PBS and then submerged in a mixed solution of 10% CCK-8 reagent with fresh medium at 37 °C for 3 h. Absorbance at 450 nm was measured using a plate reader.

In vivo experiments

Female HIGA and BALB/c mice (10 − 11 weeks) were purchased from Central Lab Animal Inc. (Seoul, Korea) (Supplementary Fig. S1A). All of the animals were maintained under specific pathogen-free conditions, and procedures were performed in accordance with an animal protocol approved by the Yeungnam University Institutional Animal Care and Use Committee (IACUC, YUMC-AEC2016-032). HIGA mice were randomly divided into 3 groups (each group, n = 25) as follow: 1) control group (HIGA), sham-operated and PBS injected; 2) naive cell group (Cell); and 3) hydrogel and cell combined group (Gel/Cell). In accordance with the recommendations of the IACUC to minimize animal sacrifice, the Gel group (without cell) was not prepared. The preparation was injected into the cortex of the left kidney by dorsal slits (length: 5 mm), the kidneys were exposed, and the right kidney was promptly removed, Supplementary Fig. S1B. The volume of injected hydrogel was 50 μL/kidney composing 35 μL gelatin and 15 μL hyaluronic acid. The combined cell number was 1 × 10⁶ cells/kidney. As a normal control, BALB/c mice were used as a background species of HIGA. Animals were sacrificed 1, 4, 8, 12, and 24 weeks after treatment, and kidney and blood were collected. The blood samples were centrifuged at 14,000 rpm for 5 min, and the clear supernatants were kept at − 80 °C until use.

Assay for renal function, histopathology, and gene expression

Serum creatinine and blood urea nitrogen (BUN) assays were used to assess renal function. The values were measured with serum obtained at each time point using a creatinine kit (R&D Systems, Minneapolis, MN, USA) and BUN kit (Roche, Basel, Switzerland) according to the manufacturer’s instructions. As a systemic biological safety index, liver function was measured (using an ELISA kit, R&D Systems) according to serum levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase (ALP).

For histopathological analysis, the kidney was fixed in 4% paraformaldehyde, and paraffin-embedded samples were cut into 5 μm sections and subjected to routine H&E staining. Renal tissues were stained with periodic acid-schiff (PAS) reagent to detect mesangial matrix expansion and segmental glomerulosclerosis. For immunohistochemistry (IHC) analysis, the sections were stained with fluorescein isothiocyanate (FITC)-labeled goat anti-mouse IgA (Life Technology, Waltham, MA, USA), and DAPI for nuclear identification. Histopathologic analysis was performed by a pathologist. The number of glomeruli were counted per unit area (4860 μm² at 100 × magnification) at week 24. For analysis of gene expression related to pro/anti-inflammation, fibrosis, and renal differentiation, the total RNA was extracted with Trizol reagent, and cDNA was synthesized with 20 μg of RNA using the Superscript Choice cDNA synthesis kit (Invitrogen, Waltham, MA, USA). For real-time PCR, the SYBR green PCR conditions were as follows: 95 °C for 10 min, followed by 40 cycles at 95 °C for 10 s, 58 °C for 50 s, and 72 °C for 20 s. The 2^−ΔΔCt method was used to analyze the relative changes in gene expression. The primer sequences, including those of the housekeeping gene (GAPDH), are listed in Table 1. For statistical analysis, a t-test and a one-way analysis of variance (ANOVA) or Tukey’s test were used. All values are expressed as the mean ± standard deviation (SD). The results were representative of at least three independent experiments.

Table 1.

Antibody and primer information for immunohistochemistry and real-time PCR

Category	Gene	Full name	Primer sequences
Pro-inflammation	IL-1ß	Interleukin-1 beta	5′-gcccatcctctgagactcat-3′
			5′-aggccacaggtattttgtcg-3′
	IL-6	Interleukin-6	5′-agttgccttcttgggactga-3′
			5′-tccacgatttcccagagaac-3′
	TNF-α	Tumor necrosis factor alpha	5′-agcccccagtctgtatcctt-3′
			5′-ctccctttgcagaactcagg-3′
Pro-fibrosis	Vimentin	Vimentin	5′-tccagatcgatgtggacgttt-3′
			5′-atactgctggcgcacatcac-3′
	Col1	Collagen type I	5′-gagcggagagtactggatcg-3′
			5′-gcttcttttccttggggttc-3′
	α-SMA	Alpha smooth muscle actin	5′-ctgacagaggcaccactgaa-3′
			5′-catctccagagtccagcaca-3′
Anti-inflammation	IL-10	Interleukin-10	5′-ccaagccttatcggaaatga-3′
			5′-ttttcacaggggagaaatcg-3′
	IL-4	Interleukin-4	5′-tcaacccccagctagttgtc-3′
			5′-tgttcttcgttgctgtgagg-3′
	IL-2	Interleukin-2	5′-cccacttcaagctccacttc-3′
			5′-atcctggggagtttcaggtt-3′
	TGF-β	Transforming growth factor beta	5′-ttgcttcagctccacagaga-3′
			5′-tggttgtagagggcaaggac-3′
Renal regeneration	Pax2	Paired box 2	5′-aaatctctatgcaaaatgacga-3′
			5′-gagagatgcagggcgatga-3′
	Wt1	Wilms tumor 1	5′-ggtccgccatcacaacatg-3′
			5′-ctttcctgcctgggatgct-3′
	Emx2	Empty spiracles homeobox 2	5′-tggccagaaagccaaagc-3′
			5′-tccgctcccaccacgtaat-3′
	vWF	von Willebrand factor	5′-gctgtgcggtgattttaacatc-3′
			5′-ccgtttacaccgctgttcct-3′
Housekeeping gene	GAPDH	Glyceraldehyde-3-phosphate dehydrogenase	5′-tgtgtccgtcgtggatctga-3′
			5′-cctgcttcaccaccttcttga-3′

Results

Characterization of CHO-HA and Gel-NH₂ hydrogel

After dialysis and freeze-drying of CHO-HA and Gel-NH₂, ¹H-NMR was used to confirm the chemical structures (Fig. 2A). ¹H-NMR(600 MHz, D₂O):(ppm) 1.9 (NH-CO-CH₃), 3.3 − 3.8 (CH-O), 4.2 − 4.4 (O–CH-O), and 4.9 − 5.1 (hemiacetalic proton formed from aldehyde groups and neighboring hydroxyl groups) was used to verify the structures of HA and CHO-HA [30, 34]. The observed resonance signals in the range of 0.8–1.5 ppm represent aliphatic protons attached to carbon atoms of valine, leucine, and isoleucine. The signals in the region of 1.5–3.0 ppm are due to the aliphatic carbon protons of arginine, leucine, lysine, proline, glycine, and aspartic acid. The peaks at 3.2 − 4.3 ppm represent the resonance signals of the α-CH of the amino acids [35, 36]. The FTIR spectra were also used to confirm the identity of CHO-HA, Gel-NH₂, and CHO-HA/Gel-NH₂ (Fig. 2B). The FTIR spectra of HA and CHO-HA were very similar due to the formation of hemiacetals. Therefore, it is hard to distinguish the signal corresponding to the aldehyde groups in the chain at 1752 cm⁻¹. The spectrum of the CHO-HA/Gel-NH₂ hydrogel shows a peak at 1650 cm⁻¹ corresponding to the –C = N group (Schiff’s base) formed between the aldehyde and amino groups, and a peak at 1520 cm⁻¹ corresponding to amide II of gelatin [21]. The intense peak at 3300 cm⁻¹ was associated with N–H stretching, overlapping with the carboxylic group’s O–H peak.

Fig. 2.

A ¹H-NMR spectra of HA, CHO-HA, and Gel-NH₂ (600 MHz, D₂O), B FTIR spectra of CHO-HA, Gel-NH₂, and CHO-HA/Gel-NH₂ hydrogel, C photograph of hydrogels and D hydrogel formation conditions at various concentrations and ratios

Hydrogel fabrication

Various concentrations of polymer precursors and ratios were evaluated for the formation of each hydrogel sample (Fig. 2C, D). The concentration of CHO-HA and Gel-NH₂ at 2% and 4% (ratio 3/7) could form a small amount of hydrogel and remain more than 50% of the solution. At a concentration 10% of both polymer precursors, hydrogel formed a total volume of solution at approximately 3–5 min. In addition, 10% CHO-HA and 20% of Gel-NH₂ showed the fastest gelation time less than 1 min and the total of polymer solution formed a stable gel. The ratio of CHO-HA/Gel-NH₂ was variously 3/7, 5/5, and 7/3 for hydrogel fabrication, and the gelation time was monitored. All compositions formed hydrogels, with a gelation time of less than 1 min, but only the hydrogel with a CHO-HA/Gel-NH₂ ratio of 3/7 maintained the gel state for 7 days. These hydrogel systems exhibited injectability, as shown in Supplementary Fig. S2A.

Because native gelatin is not soluble in water at room temperature, it was not employed to fabricate the hydrogel in this work. However, when the temperature is raised above 37 °C, gelatin becomes soluble. Because of intermolecular physical cross-linkages, the gelatin solution returns to a gel-like state when it cools to ambient temperature. Gel-NH₂ overcomes this problem because Gel-NH₂ could be used at room temperature in a solution state. The introduction of amine groups reduced the intrinsic viscosity of gelatin, owing to the intramolecular attraction between the negatively charged carboxyl groups and the positively charged amino groups. Moreover, the hydrogel prepared from native gelatin and CHO-HA was unstable when the temperature raised to 37 °C, (Supplementary Fig. S2B).

Rheological properties

The rheological properties of the hydrogels were measured by monitoring the storage modulus (G’) as a function of the angular frequency at 25 °C (Fig. 3). The storage modulus gradually increased with increasing frequency. The hydrogel with a CHO-HA/Gel-NH₂ ratio of 3/7 exhibited the highest G’ ($\approx$ 1.75 × 10³ Pa; Fig. 3A), compared with the CHO-HA/Gel-NH₂ system with a ratio of 5/5 (G’ $\approx$ 1.2 × 10³ Pa), and CHO-HA/Gel-NH₂ with a ratio of 7/3 (G’ $\approx$ 0.3 × 10³ Pa; Fig. 3B, C).

Fig. 3.

A–C Frequency dependence of storage modulus (G′) for hydrogels with different CHO-HA/Gel-NH₂ ratios: A 3/7, B 5/5, and C 7/3. D–I SEM images of cross-sections of CHO-HA/Gel-NH₂ hydrogels, D ratio 3/7 (scale 500 μm, 5 kV), E ratio 5/5 (scale 500 μm, 5 kV), F ratio 7/3 (scale 500 μm, 5 kV). CHO-HA/Gel-NH₂ hydrogel with 3/7 ratio after seven days: G scale 500 μm, 5 kV, H scale 200 μm, 5 kV, and I scale 50 μm, 5 kV. J Swelling ratio of CHO-HA/Gel-NH₂ hydrogel with 3/7, 5/5 and 7/3 ratio and K Swelling ratio and degradation of CHO-HA/Gel-NH₂ hydrogel with 3/7 continued to seven days

Morphology of hydrogel

The cross-sectional SEM images of the lyophilized CHO-HA/Gel-NH₂-based hydrogels are shown in Fig. 3D–I. The hydrogel at a ratio 3/7 exhibited a homogeneous distribution of interconnected porous structures with a well-defined pore size (~ 200 μm). The matrix surrounding the hydrogel’s pores had a smooth surface (Fig. 3D). The hydrogels with CHO-HA/Gel-NH₂ ratios of 5/5 and 7/3 did not possess well-defined porous structures, and the pore size appeared larger for the hydrogel with the 5/5 ratio (Fig. 3E). The pore structure was non-homogeneous for the hydrogel with the 7/3 ratio (Fig. 3F). Figure 3G–I show the morphology of the degradable hydrogel at a ratio of 3/7 over 7 days after freeze-drying. The hydrogel degraded, and it can be seen by porous structure of the hydrogel collapsed.

Equilibrium swelling ratio and degradation of hydrogel

The equilibrium swelling ratio of the hydrogels were measured at pH 7.4 and 37 °C for 24 h, and the hydrogel degradation was subsequently evaluated over 7 days. The PBS uptake capacity determines the water absorption capacity of the hydrogel, which is an important factor affecting the biological activity. The swelling ratio of the CHO-HA/Gel-NH₂ (ratio: 3/7) hydrogel increased sharply during the first 12 h and decreased continuously to 7 days (Fig. 3J, K). On the other hand, after incubation, the swelling ratio of the hydrogels with CHO-HA/Gel-NH₂ ratios of 5/5 and 7/3 decreased dramatically in the first 2 h; however, they could not be further analyzed because of degradation of hydrogel. Thus, cross-linking of CHO-HA/Gel-NH₂ at a 3/7 ratio was suitable for preparing the hydrogel.

In vitro cell viability and cell proliferation

The cytotoxicity of the CHO-HA/Gel-NH₂ hydrogels was first investigated in vitro using a live/dead assay (Fig. 4A). Cell proliferation was further evaluated using CCK-8 assay (Fig. 4B). It was observed that the hydrogel promoted the cell viability and proliferation in the incubation of the iNPCs. The proliferative capacity of the cells seeded on the hydrogel and encapsulated in the hydrogel increased steadily from day 1 to day 7 as measured by CCK-8 assay. The number of cells in Matrigel increased sharply from day 1 to day 3, being approximately twice that of the cells incubated with hydrogels. Then, the viability of the cells cultured on Matrigel declined from day 3 to day 7, which could be described as limited oxygen and metabolite diffusion with increasing cell density in the 2D culture on the plates [37]. Moreover, cells growth increased rapidly until day 3 and became confluent, resulting in cells detached from the surface of cell culture plate. Interestingly, there was a marked increase in the cell number after 3 days of cell culture in culture systems employing the hydrogel.

Fig. 4.

A Live/dead cell assay observed by fluorescence microscopy. Live cells emit green fluorescence and dead cells emit red fluorescence (red arrows). Scale bar 500 μm, magnification × 10. B CCK-8 assay of cells seeded on Matrigel coated on culture plate, cells seeded on CHO-HA/Gel-NH₂ hydrogel, and cells encapsulated in CHO-HA/Gel-NH₂ hydrogel. For statistical analysis, a two-way analysis of variance (ANOVA) or Tukey’s test were used. All values are expressed as the mean ± standard deviation (SD). Results are representative of at least three experiments. Statistical significance: ****p < 0.0001, **p < 0.01

In vivo experiments

The gross images of the kidneys of mice with IgA nephropathy showed that the kidneys of the Cell and Gel/Cell group had a pale color compared to the controls at week 1, with the color becoming normal at week 4 (Supplementary, Fig. S1C). The color change is considered a temporary occurrence caused by an injection of the cells and gel into the renal cortex. During the 24-weeks, the experimental groups did not exhibit a significant difference in body weight (Supplementary Fig. S3). The food intake was almost the same among these groups (5.0–7.5 g per day per mouse, data not shown). The renal histological changes in response to treatment with the hydrogel combination were identified by H&E staining (Fig. 5). The HIGA mice showed severe renal interstitial area necrosis (arrow), mesangial expansion (arrow head), and segmental glomerulosclerosis (asterisk) from week 8. The Cell group showed mild mesangial proliferation, less mesangial matrix expansion, and fewer fibrocellular crescents. The Gel/Cell group showed an almost normal appearance (blue arrow, injected hydrogel). Mesangial depositions of IgA were examined by immunofluorescence staining. The observed IgA deposits in the HIGA mice were diffusely and globally located in the mesangial area, and the appearance was frequently condensed, whereas mesangial deposition of IgA was weak in the Cell group at week 1 and 4, and from week 8, IgA was noticeably increased. Gel/Cell group showed consistent inhibition of IgA deposition until week 24 (Supplementary Fig. S4). To examine whether these histological improvements affected the renal function, the BUN and serum creatinine levels were assessed (Fig. 6). The BUN and serum creatinine scores of the Gel/Cell group were statistically lower than those of the Cell group from week 4 (p < 0.05). This statistical comparison was performed only between the Cell and Gel/Cell groups because the in vivo study focused on the enhanced therapeutic effects of the hydrogel combination compared to naive cell therapy. There were no statistical differences in the analysis of serum AST, ALT, and ALP levels, indicating that the kidney treatments had no effect to liver function (Supplementary Fig. S5).

Fig. 5.

Histopathology with H&E staining for detection of mesangial proliferation and glomerular sclerosis, (magnification × 400). Experiments were divided into 4 groups (each group n = 25), 1: BALB/c (background species of HIGA), 2: HIGA (sham operated and PBS injected HIGA mice), 3: Cell (naive cell injected HIGA mice group) and 4: Gel/Cell (hydrogel and cell injected HIGA mice group). Scale bars 100 μm

Fig. 6.

Blood urea nitrogen (BUN) and serum creatinine values for analysis of renal function. For statistical analysis, a t-test and a one-way analysis of variance (ANOVA) or Tukey’s test were used. Data are presented as means ± SD. *p < 0.05, **p < 0.01, Gel/Cell versus Cell group

The molecular mechanisms were investigated because the Gel/Cell combination improved the kidney structure and function. The expression of pro-inflammatory and pro-fibrotic molecules, such as IL-1β, IL-6, TNF-α, vimentin, Col-1, and α-SMA, was significantly reduced in the Gel/Cell group compared to the Cell group (Fig. 7). The expression of anti-inflammatory genes was enhanced in the Gel/Cell group compared to the Cell group (Fig. 8). These anti-inflammatory and anti-fibrotic effects were stimulated by renal differentiation related to gene expression. The Gel/Cell group showed enhanced Pax2, Wt1, Emx2, and vWf expression than the Cell group (Fig. 9).

Fig. 7.

Expression of pro-inflammatory genes (IL-1β, IL-6 and TNF-α) and pro-fibrosis genes (Vimentin, Col 1 and α-SMA). For statistical analysis, a t-test and a one-way analysis of variance (ANOVA) or Tukey’s test were used. Data are presented as means ± SD. *p < 0.05, **p < 0.01, Gel/Cell versus Cell group

Fig. 8.

Expression of anti-inflammatory genes (IL-10, IL-4, IL-2 and TGF-β). For statistical analysis, a t-test and a one-way analysis of variance (ANOVA) or Tukey’s test were used. Data are presented as means ± SD. *p < 0.05, **p < 0.01, Gel/Cell versus Cell group

Fig. 9.

Expression of renal differentiation genes (Pax2, Wt1, Emx2 and vWF) related to renal differentiation. For statistical analysis, a t-test and a one-way analysis of variance (ANOVA) or Tukey’s test were used. Data are presented as means ± SD. *p < 0.05, **p < 0.01, Gel/Cell versus Cell groups

Discussion

CHO-HA/Gel-NH₂ was created via Schiff-base reaction without additional cross-linking agents. Hydrogel at ratio 3/7 exhibited the greatest mechanical properties among three different ratios. These results are attributed to the effects on the mechanical properties of the hydrogel of increasing the cross-linking density. For each composite ratio of hydrogel, the G’ was observed to be greater than the G’’ over the entire range of frequency, indicating a viscoelastic behavior with a dominant elastic properties. The greatest G’ value was observed for the composite hydrogel at a ratio 3/7, perhaps resulting from the greatest extent of Schiff’s-base cross-linking [30]. The G’ of the hydrogel enhanced when the cross-linking density increased, and more stable hydrogels were typically constructed. This is probable due to a high degree of intramolecular cross-linking at high cross-linking density [38]. Thus, cross-linking density can be used to control the mechanical properties of the hydrogel [39]. CHO-HA and Gel-NH₂ are macromolecular with long polymer chains and the cross-linking density is low, result in a release cross-linking network. According to the results, the hydrogel with a CHO-HA/Gel-NH₂ ratio of 3/7 has an appropriate cross-linking density, with cross-links acting as tethering points between different hydrogel’s various chains, preventing dissociation.

The microstructure of the hydrogel plays an important role in improving the mechanical properties. The hydrogel with a CHO-HA/Gel-NH₂ ratio of 3/7 displayed being interconnected and mutually penetrating with pore size 200 μm. These results are supported by A. Sarker et al. [39], wherein the pores are approximately 100–250 μm in size, which is a critical range for tissue regeneration. We deduced that this ratio might have a high permeability for nutrients and support cellular growth [34].

A high degree of swelling is a requirement for the 3D scaffolds to serve as a matrix for tissue regeneration. This is due to the solution’s wicking activity through the hydrogel’s pores, as well as the hydration of free -OH groups in the system. This free diffusion does not seem to change the morphology of the hydrogel extensively. The highest water uptake ability of the freeze-dried hydrogel, even within 2 h, suggests the potential ability of the hydrogel to absorb and supply nutrients to all the cells that are seeded or encapsulated within the porous structure of the hydrogel [40, 41]. This may also facilitate the diffusion of the cells into the interior of the hydrogel while seeding.

iNPCs were generated from UCs. At 12 days post-transduction, cell colonies were observed and expanded as single cells on Matrigel-coated plates. Upon further passaging, putative iNPCs exhibiting a small, spindle-shaped mesenchymal morphology were observed. Consequently, cellular proliferation was investigated by culturing iNPCs with hydrogels. Porous hydrogels play a crucial role in cellular growth and are necessary for tissue regeneration.

Moreover, the porosity of the hydrogel is important for nutrient exchange and diffusion across the material. However, when the cells were encapsulated, the hydrogel disintegrated within 24 h of incubation, resulting in decreasing cell viability. This is due to the intrinsic volume of cells in gels lowering the mechanical characteristics of the hydrogel. Moreover, a significant amount of unreacted aldehyde groups and amino groups may have remained, resulting in loss of cell viability on both hydrogel culture systems. This may be due to the viscous reaction medium, which prevented chains from coming in contact to link together by covalent cross-linking [42]. However, iNPCs retained their proliferative capacity in the hydrogel microenvironments over incubation time. These results suggest that the hydrogel has excellent biocompatibility, which may be beneficial for the treatment of IgAN.

According to the swelling ratio, degradation, and biocompatibility in vitro, hydrogel rapidly degraded in culture media due to the high back-and-forth reaction rate at natural pH. Moreover, the hydrogel degraded faster under acidic conditions via hydrolysis reaction due to the limitation of cell culture in vitro. However, the Schiff-base reaction naturally exists in the human body and it plays an important role in self-healing, so an in vivo study was conducted to investigate prolonging the therapeutic potential.

H&E and PAS staining were used to identify the renal histological changes in response to treatment with URPC and hydrogel combination. The hydrogel combined with cells showed better mesangial proliferation and less immune deposition compared with cells group. BUN and serum creatinine levels were measured in HIGA mice to evaluate whether these histological improvements affected on renal function. BUN is urea nitrogen derived from the breakdown of protein, and serum creatinine is a waste product of muscle degradation, both of which are filtered from blood by the glomeruli. The normal BUN rang is 6.0–23 mg/dL in mice [43, 44]. At 12 weeks to 24 weeks in HIGA mice, the BUN level increased to 35 mg/dL, higher than the normal range. However, Gel/Cell group showed a lower BUN value than the Cell group and Control group at 24 weeks. Likewise, serum creatinine value in Gel/Cell group became less at 8 weeks and remained steady at 3.24 mg/dL until 24 weeks. Thus, it was observed that BUN and serum creatinine decreased along with the improvement of renal pathological findings in Cells/Gel group comparing with HIGA and Cell groups. To determine the pro-inflammatory, pro-fibrosis and anti-inflammatory effects of CHO-HA/Gel-NH₂ hydrogel, we analyzed gene expression using RT-PCR. These data indicated that the pro-inflammatory and pro-fibrosis actions of URPC could be strengthened by hydrogel.

According to gene expression analysis, URPC administrated hydrogel reduced the expression of pro-inflammatory (IL-1β, IL-6 and TNF-α) and pro-fibrosis genes (Vimentin, Col-1 and α-SMA) [45, 46]. TNF-α is expressed in inflammation sites and injury, which subsequently causes the release of inflammatory mediators and other pro-inflammatory cytokines [47]. It has also been reported that TNF-α activates caspase-mediated apoptosis through activation of the death receptor pathway. IL-6 is expressed mainly in mesangial cells, and its high concentration promotes chronic renal disease due to abnormal permeability of glomerular endothelium, fibronectin overexpression and mesangial cell expansion [48]. In IgAN pathogenesis, ECM expansion or mesangial proliferation are frequently corresponded to serum elevation of IgA-IgG2a IC and strengthened IgA deposition, as well as cytokine production such as IL-6 and TGF-β [49]. Moreover, TNF-α and IL-6 also remarkably increase in patients with chronic renal failure. It has been suggested that myofibroblasts participate in the healing of mechanical wounds. Myofibroblasts that synthesize and secrete Col-1 during wound healing are characterized by the expression of smooth muscle (α-SMA) [50]. The present study further observed significant differences in the expression of anti-inflammatory, including IL-10, TGF-β, IL-4 and IL-2, in kidneys from URPC/Gel-treated HIGA mice compared with control mice. It has been reported that deficiency in TGF- expression may increase the severity of the renal injury, as indicated by more severe renal tubular damage and increased serum creatinine and BUN levels. These findings result in an increased risk of renal ischemia/reperfusion and inducing kidney injury. In addition, inhibition of TGF-β has been shown to facilitate fibrosis in most models of experimental CKD. TGF-β induces ‘pro-fibrotic’ responses in numerous cell types, including increased ECM synthesis by mesangial cells, renal fibroblasts, and chemokine release [51]. By contrast, overexpression of TGF-β in transgenic mice causes renal fibrosis. In addition, increased IL-10 production may protect animals from renal injury [52]. Thus, our results suggest that administering URPC in hydrogel may improve functional parameters and reduce the progression of renal fibrosis during the early and late stages following injury, in association with the promotion of IL-10, IL-4, IL-2 and TGF-β, and the inhibition of IL-1β, IL-6, TNF-α, Vimentin, Col-1 and α-SMA. Expression of genes related to renal differentiation (Pax2, Wt1, Emx2 and vWF) supported that this present hydrogel could enhance renal regeneration. Some research groups reported the existence of progenitor cells in the kidney and demonstrated the presence of a resident population of PAX-2 expression in adult normal human kidneys, suggesting that these cells are capable of expansion and self-renewal [53].

The therapeutic effects of stem cells (intrinsic or extrinsic) are due to their multiple mechanisms, such as immunomodulation and differentiation into organ-specific cells. In addition, the hydrogel provides an optimum environment for stem cell function [54]. The hydrogel used in this study was composed of gelatin and HA. Gelatin can easily embed cells, and HA can increase the stiffness of the matrix [24]. This artificial hydrogel can mimic natural ECM in terms of water retention and nutrients and metabolite permeation, mechanical properties, and biocompatibility [55]. These properties provide an appropriate microenvironment for cell survival, adhesion, proliferation, differentiation, and release of bioactive molecules. Furthermore, the hydrogel is injectable, so that cells can be introduced directly into the injured area, overcoming the problem of cell diffusion and enabling the sustained release of paracrine molecules [56]. Therefore, our findings suggest that our developed hydrogels reduce IgAN more effectively than naive cell therapy. This simple approach may become a preferred treatment for renal failure patients.

Supplementary Information

Below is the link to the electronic supplementary material.

Declarations

Conflict of interest

The authors declared that they have no conflicts of interest to this work.

Ethical statement

The animal studies were performed procedures following an animal protocol approved by the Yeungnam University Institutional Animal Care and Use Committee (IACUC, YUMC-AEC2016-032).